scvi.dataloaders.DataSplitter#

class scvi.dataloaders.DataSplitter(adata_manager, train_size=0.9, validation_size=None, use_gpu=False, **kwargs)[source]#

Creates data loaders train_set, validation_set, test_set.

If train_size + validation_set < 1 then test_set is non-empty.

Parameters:
  • adata_manager (AnnDataManager) – AnnDataManager object that has been created via setup_anndata.

  • train_size (float (default: 0.9)) – float, or None (default is 0.9)

  • validation_size (Optional[float] (default: None)) – float, or None (default is None)

  • use_gpu (bool (default: False)) – Use default GPU if available (if None or True), or index of GPU to use (if int), or name of GPU (if str, e.g., ‘cuda:0’), or use CPU (if False).

  • **kwargs – Keyword args for data loader. If adata has labeled data, data loader class is SemiSupervisedDataLoader, else data loader class is AnnDataLoader.

Examples

>>> adata = scvi.data.synthetic_iid()
>>> scvi.model.SCVI.setup_anndata(adata)
>>> adata_manager = scvi.model.SCVI(adata).adata_manager
>>> splitter = DataSplitter(adata)
>>> splitter.setup()
>>> train_dl = splitter.train_dataloader()

Attributes table#

CHECKPOINT_HYPER_PARAMS_KEY

CHECKPOINT_HYPER_PARAMS_NAME

CHECKPOINT_HYPER_PARAMS_TYPE

hparams

The collection of hyperparameters saved with save_hyperparameters().

hparams_initial

The collection of hyperparameters saved with save_hyperparameters().

name

Methods table#

add_argparse_args(parent_parser, **kwargs)

Extends existing argparse by default LightningDataModule attributes.

from_argparse_args(args, **kwargs)

Create an instance from CLI arguments.

from_datasets([train_dataset, val_dataset, ...])

Create an instance from torch.utils.data.Dataset.

get_init_arguments_and_types()

Scans the DataModule signature and returns argument names, types and default values.

load_from_checkpoint(checkpoint_path[, ...])

Primary way of loading a datamodule from a checkpoint.

load_state_dict(state_dict)

Called when loading a checkpoint, implement to reload datamodule state given datamodule state_dict.

on_after_batch_transfer(batch, dataloader_idx)

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

on_before_batch_transfer(batch, dataloader_idx)

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

on_load_checkpoint(checkpoint)

Called by Lightning to restore your model.

on_save_checkpoint(checkpoint)

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

predict_dataloader()

Implement one or multiple PyTorch DataLoaders for prediction.

prepare_data()

Use this to download and prepare data.

save_hyperparameters(*args[, ignore, frame, ...])

Save arguments to hparams attribute.

setup([stage])

Split indices in train/test/val sets.

state_dict()

Called when saving a checkpoint, implement to generate and save datamodule state.

teardown([stage])

Called at the end of fit (train + validate), validate, test, or predict.

test_dataloader()

Create test data loader.

train_dataloader()

Create train data loader.

transfer_batch_to_device(batch, device, ...)

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

val_dataloader()

Create validation data loader.

Attributes#

CHECKPOINT_HYPER_PARAMS_KEY

DataSplitter.CHECKPOINT_HYPER_PARAMS_KEY = 'datamodule_hyper_parameters'#

CHECKPOINT_HYPER_PARAMS_NAME

DataSplitter.CHECKPOINT_HYPER_PARAMS_NAME = 'datamodule_hparams_name'#

CHECKPOINT_HYPER_PARAMS_TYPE

DataSplitter.CHECKPOINT_HYPER_PARAMS_TYPE = 'datamodule_hparams_type'#

hparams

DataSplitter.hparams[source]#

The collection of hyperparameters saved with save_hyperparameters(). It is mutable by the user. For the frozen set of initial hyperparameters, use hparams_initial.

Return type:

Union[AttributeDict, MutableMapping]

Returns:

Mutable hyperparameters dictionary

hparams_initial

DataSplitter.hparams_initial[source]#

The collection of hyperparameters saved with save_hyperparameters(). These contents are read-only. Manual updates to the saved hyperparameters can instead be performed through hparams.

Returns:

immutable initial hyperparameters

Return type:

AttributeDict

name

DataSplitter.name: str = Ellipsis#

Methods#

add_argparse_args

classmethod DataSplitter.add_argparse_args(parent_parser, **kwargs)[source]#

Extends existing argparse by default LightningDataModule attributes.

Example:

parser = ArgumentParser(add_help=False)
parser = LightningDataModule.add_argparse_args(parser)
Return type:

ArgumentParser

from_argparse_args

classmethod DataSplitter.from_argparse_args(args, **kwargs)[source]#

Create an instance from CLI arguments.

Parameters:
  • args (Union[Namespace, ArgumentParser]) – The parser or namespace to take arguments from. Only known arguments will be parsed and passed to the LightningDataModule.

  • **kwargs – Additional keyword arguments that may override ones in the parser or namespace. These must be valid DataModule arguments.

Example:

module = LightningDataModule.from_argparse_args(args)

from_datasets

classmethod DataSplitter.from_datasets(train_dataset=None, val_dataset=None, test_dataset=None, predict_dataset=None, batch_size=1, num_workers=0)[source]#

Create an instance from torch.utils.data.Dataset.

Parameters:
  • train_dataset (Union[Dataset, Sequence[Dataset], Mapping[str, Dataset], None] (default: None)) – (optional) Dataset to be used for train_dataloader()

  • val_dataset (Union[Dataset, Sequence[Dataset], None] (default: None)) – (optional) Dataset or list of Dataset to be used for val_dataloader()

  • test_dataset (Union[Dataset, Sequence[Dataset], None] (default: None)) – (optional) Dataset or list of Dataset to be used for test_dataloader()

  • predict_dataset (Union[Dataset, Sequence[Dataset], None] (default: None)) – (optional) Dataset or list of Dataset to be used for predict_dataloader()

  • batch_size (int (default: 1)) – Batch size to use for each dataloader. Default is 1.

  • num_workers (int (default: 0)) – Number of subprocesses to use for data loading. 0 means that the data will be loaded in the main process. Number of CPUs available.

get_init_arguments_and_types

classmethod DataSplitter.get_init_arguments_and_types()[source]#

Scans the DataModule signature and returns argument names, types and default values.

Returns:

(argument name, set with argument types, argument default value).

Return type:

List with tuples of 3 values

load_from_checkpoint

classmethod DataSplitter.load_from_checkpoint(checkpoint_path, hparams_file=None, **kwargs)[source]#

Primary way of loading a datamodule from a checkpoint. When Lightning saves a checkpoint it stores the arguments passed to __init__ in the checkpoint under "datamodule_hyper_parameters".

Any arguments specified through **kwargs will override args stored in "datamodule_hyper_parameters".

Parameters:
  • checkpoint_path (Union[str, Path, IO]) – Path to checkpoint. This can also be a URL, or file-like object

  • hparams_file (Union[str, Path, None] (default: None)) –

    Optional path to a .yaml or .csv file with hierarchical structure as in this example:

    dataloader:
        batch_size: 32
    

    You most likely won’t need this since Lightning will always save the hyperparameters to the checkpoint. However, if your checkpoint weights don’t have the hyperparameters saved, use this method to pass in a .yaml file with the hparams you’d like to use. These will be converted into a dict and passed into your LightningDataModule for use.

    If your datamodule’s hparams argument is Namespace and .yaml file has hierarchical structure, you need to refactor your datamodule to treat hparams as dict.

  • **kwargs – Any extra keyword args needed to init the datamodule. Can also be used to override saved hyperparameter values.

Returns:

LightningDataModule instance with loaded weights and hyperparameters (if available).

Note

load_from_checkpoint is a class method. You should use your LightningDataModule class to call it instead of the LightningDataModule instance.

Example:

# load weights without mapping ...
datamodule = MyLightningDataModule.load_from_checkpoint('path/to/checkpoint.ckpt')

# or load weights and hyperparameters from separate files.
datamodule = MyLightningDataModule.load_from_checkpoint(
    'path/to/checkpoint.ckpt',
    hparams_file='/path/to/hparams_file.yaml'
)

# override some of the params with new values
datamodule = MyLightningDataModule.load_from_checkpoint(
    PATH,
    batch_size=32,
    num_workers=10,
)

load_state_dict

DataSplitter.load_state_dict(state_dict)[source]#

Called when loading a checkpoint, implement to reload datamodule state given datamodule state_dict.

Parameters:

state_dict (Dict[str, Any]) – the datamodule state returned by state_dict.

Return type:

None

on_after_batch_transfer

DataSplitter.on_after_batch_transfer(batch, dataloader_idx)[source]#

Override to alter or apply batch augmentations to your batch after it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters:
  • batch (Any) – A batch of data that needs to be altered or augmented.

  • dataloader_idx (int) – The index of the dataloader to which the batch belongs.

Return type:

Any

Returns:

A batch of data

Example:

def on_after_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = gpu_transforms(batch['x'])
    return batch
Raises:

MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

on_before_batch_transfer

DataSplitter.on_before_batch_transfer(batch, dataloader_idx)[source]#

Override to alter or apply batch augmentations to your batch before it is transferred to the device.

Note

To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters:
  • batch (Any) – A batch of data that needs to be altered or augmented.

  • dataloader_idx (int) – The index of the dataloader to which the batch belongs.

Return type:

Any

Returns:

A batch of data

Example:

def on_before_batch_transfer(self, batch, dataloader_idx):
    batch['x'] = transforms(batch['x'])
    return batch
Raises:

MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

on_load_checkpoint

DataSplitter.on_load_checkpoint(checkpoint)[source]#

Called by Lightning to restore your model. If you saved something with on_save_checkpoint() this is your chance to restore this.

Parameters:

checkpoint (Dict[str, Any]) – Loaded checkpoint

Example:

def on_load_checkpoint(self, checkpoint):
    # 99% of the time you don't need to implement this method
    self.something_cool_i_want_to_save = checkpoint['something_cool_i_want_to_save']

Note

Lightning auto-restores global step, epoch, and train state including amp scaling. There is no need for you to restore anything regarding training.

Return type:

None

on_save_checkpoint

DataSplitter.on_save_checkpoint(checkpoint)[source]#

Called by Lightning when saving a checkpoint to give you a chance to store anything else you might want to save.

Parameters:

checkpoint (Dict[str, Any]) – The full checkpoint dictionary before it gets dumped to a file. Implementations of this hook can insert additional data into this dictionary.

Example:

def on_save_checkpoint(self, checkpoint):
    # 99% of use cases you don't need to implement this method
    checkpoint['something_cool_i_want_to_save'] = my_cool_pickable_object

Note

Lightning saves all aspects of training (epoch, global step, etc…) including amp scaling. There is no need for you to store anything about training.

Return type:

None

predict_dataloader

DataSplitter.predict_dataloader()[source]#

Implement one or multiple PyTorch DataLoaders for prediction.

It’s recommended that all data downloads and preparation happen in prepare_data().

Note

Lightning adds the correct sampler for distributed and arbitrary hardware There is no need to set it yourself.

Return type:

Union[DataLoader, Sequence[DataLoader]]

Returns:

A torch.utils.data.DataLoader or a sequence of them specifying prediction samples.

Note

In the case where you return multiple prediction dataloaders, the predict_step() will have an argument dataloader_idx which matches the order here.

prepare_data

DataSplitter.prepare_data()[source]#

Use this to download and prepare data. Downloading and saving data with multiple processes (distributed settings) will result in corrupted data. Lightning ensures this method is called only within a single process, so you can safely add your downloading logic within.

Warning

DO NOT set state to the model (use setup instead) since this is NOT called on every device

Example:

def prepare_data(self):
    # good
    download_data()
    tokenize()
    etc()

    # bad
    self.split = data_split
    self.some_state = some_other_state()

In a distributed environment, prepare_data can be called in two ways (using prepare_data_per_node)

  1. Once per node. This is the default and is only called on LOCAL_RANK=0.

  2. Once in total. Only called on GLOBAL_RANK=0.

Example:

# DEFAULT
# called once per node on LOCAL_RANK=0 of that node
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = True


# call on GLOBAL_RANK=0 (great for shared file systems)
class LitDataModule(LightningDataModule):
    def __init__(self):
        super().__init__()
        self.prepare_data_per_node = False

This is called before requesting the dataloaders:

model.prepare_data()
initialize_distributed()
model.setup(stage)
model.train_dataloader()
model.val_dataloader()
model.test_dataloader()
model.predict_dataloader()
Return type:

None

save_hyperparameters

DataSplitter.save_hyperparameters(*args, ignore=None, frame=None, logger=True)[source]#

Save arguments to hparams attribute.

Parameters:
  • args (Any) – single object of dict, NameSpace or OmegaConf or string names or arguments from class __init__

  • ignore (Union[Sequence[str], str, None] (default: None)) – an argument name or a list of argument names from class __init__ to be ignored

  • frame (Optional[frame] (default: None)) – a frame object. Default is None

  • logger (bool (default: True)) – Whether to send the hyperparameters to the logger. Default: True

Example::
>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # manually assign arguments
...         self.save_hyperparameters('arg1', 'arg3')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14
>>> class AutomaticArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # equivalent automatic
...         self.save_hyperparameters()
...     def forward(self, *args, **kwargs):
...         ...
>>> model = AutomaticArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg2": abc
"arg3": 3.14
>>> class SingleArgModel(HyperparametersMixin):
...     def __init__(self, params):
...         super().__init__()
...         # manually assign single argument
...         self.save_hyperparameters(params)
...     def forward(self, *args, **kwargs):
...         ...
>>> model = SingleArgModel(Namespace(p1=1, p2='abc', p3=3.14))
>>> model.hparams
"p1": 1
"p2": abc
"p3": 3.14
>>> class ManuallyArgsModel(HyperparametersMixin):
...     def __init__(self, arg1, arg2, arg3):
...         super().__init__()
...         # pass argument(s) to ignore as a string or in a list
...         self.save_hyperparameters(ignore='arg2')
...     def forward(self, *args, **kwargs):
...         ...
>>> model = ManuallyArgsModel(1, 'abc', 3.14)
>>> model.hparams
"arg1": 1
"arg3": 3.14
Return type:

None

setup

DataSplitter.setup(stage=None)[source]#

Split indices in train/test/val sets.

state_dict

DataSplitter.state_dict()[source]#

Called when saving a checkpoint, implement to generate and save datamodule state.

Return type:

Dict[str, Any]

Returns:

A dictionary containing datamodule state.

teardown

DataSplitter.teardown(stage=None)[source]#

Called at the end of fit (train + validate), validate, test, or predict.

Parameters:

stage (Optional[str] (default: None)) – either 'fit', 'validate', 'test', or 'predict'

Return type:

None

test_dataloader

DataSplitter.test_dataloader()[source]#

Create test data loader.

train_dataloader

DataSplitter.train_dataloader()[source]#

Create train data loader.

transfer_batch_to_device

DataSplitter.transfer_batch_to_device(batch, device, dataloader_idx)[source]#

Override this hook if your DataLoader returns tensors wrapped in a custom data structure.

The data types listed below (and any arbitrary nesting of them) are supported out of the box:

For anything else, you need to define how the data is moved to the target device (CPU, GPU, TPU, …).

Note

This hook should only transfer the data and not modify it, nor should it move the data to any other device than the one passed in as argument (unless you know what you are doing). To check the current state of execution of this hook you can use self.trainer.training/testing/validating/predicting so that you can add different logic as per your requirement.

Note

This hook only runs on single GPU training and DDP (no data-parallel). Data-Parallel support will come in near future.

Parameters:
  • batch (Any) – A batch of data that needs to be transferred to a new device.

  • device (device) – The target device as defined in PyTorch.

  • dataloader_idx (int) – The index of the dataloader to which the batch belongs.

Return type:

Any

Returns:

A reference to the data on the new device.

Example:

def transfer_batch_to_device(self, batch, device, dataloader_idx):
    if isinstance(batch, CustomBatch):
        # move all tensors in your custom data structure to the device
        batch.samples = batch.samples.to(device)
        batch.targets = batch.targets.to(device)
    elif dataloader_idx == 0:
        # skip device transfer for the first dataloader or anything you wish
        pass
    else:
        batch = super().transfer_batch_to_device(data, device, dataloader_idx)
    return batch
Raises:

MisconfigurationException – If using data-parallel, Trainer(strategy='dp').

See also

  • move_data_to_device()

  • apply_to_collection()

val_dataloader

DataSplitter.val_dataloader()[source]#

Create validation data loader.